Have you tested the large ring with two the two pulley set up? Like you did the smaller ring. The small one looks really smooth as it tracks across.
The smaller one is noticeably smoother (and cheaper to make/ship) so I’ve pretty much committed to the smaller one unless I can find a reason to go back to the big one.
I need to get all the kinematics working so I can really test cutting and measuring shapes before we’ll have a real verdict.
3 Likes
Is it perhaps that the smaller one is smoother due to the pulley spacing being more optimal for the arc radius?
1 Like
I believe it has to do with the smaller ring having more mechanical advantage. The large ring acts like a long lever amplifying the friction in the pulleys
3 Likes
I have to add my 50c:
I do have concerns with that design. I think that friction makes this all a bit imprecise in the end (this can be easily seen on the videos above - the pulleys are “jumping” rather than moving smoothly).
Geometrically this for sure introduces an error.
I guess the final construction might rather be, to connect the chains and let them run around the router (the chain-loop obviously needs to be fixated to the router) - it should be easy to construct this accordingly.
That way it is ensured, that the bit always has the same distance to the chain-loop and this finally results in the sought-after mathematical stability - the bit is just not in the very corner of the true triangle that the chains form, but in the corner of a “virtual” triangle with the bit in its center.
I might have time tomorrow to draw that schematically.
I have to add my 50c:
I do have concerns with that design. I think that friction makes this all a bit imprecise in the end (this can be easily seen on the videos above - the pulleys are “jumping” rather than moving smoothly).
Geometrically this for sure introduces an error.
how much error, especially compared to other options. Remember that the video
was of a setup thrown together from very cheap pullies.
I guess the final construction might rather be, to connect the chains and let them run around the router (the chain-loop obviously needs to be fixated to the router) - it should be easy to construct this accordingly.
That way it is ensured, that the bit always has the same distance to the chain-loop and this finally results in the sought-after mathematical stability.
I’m not visualizing what you are suggesting.
For what it’s worth I only live a few miles from Sailrite and literally drive past it twice a day.
If you want me to pick up a few items for testing I’d be happy to do so.
So, for example, they have stainless steel “hoops”: http://www.sailrite.com/All-Hardware/Sail-Hardware/Other-Sail-Hardware
and a menagerie of blocks: http://www.sailrite.com/All-Hardware/Sail-Hardware/Blocks-and-Fairleads
I also pass 80/20 every day (they’re only a few miles from Sailrite). They have a dizzying array of things that could be used on Maslow.
I had no idea they sold metal rings like those! Cool!
So far I’ve learned that the smaller the circle the better (I removed my router handles to get it tighter)
The jerky movements bug has my full attention right now, but as soon as we’ve got it tracked down I’m going to be able to spend more time on this and I would love a second set of opinions if you’ve got the time to put together something similar and compare results
2 Likes
Hey Bar,
Love your leadership and the constant improvement and contribution from the community.
Was wondering if you might want to test these readily available options?
http://tinyurl.com/yd736zmq
It seems structurally sound if we just raise the z-depth for the chains.
What appealed was that these come with a centered opening about right for the router.
Cheers!
Tom
1 Like
I was thinking the same thing at first, then I realized that a lazy susan would only freely rotate with one point. To keep the bit in the center we need both chain attachment points to independently rotate around the bit.
Can we use these Kung Fu rings? They are very strong, won’t bend and are cheap:
http://s.aliexpress.com/bU77FRzq
1 Like
I sure did not want to offend you by referring to your video,
I just tried to think about a solution to a possible issue.
I have created a schematic drawing of what I was trying to explain.
The wood-ring-plate could as well just be a half-circle,
since the chains would only just run around the lower arc.
Sure some construction is needed to fit such a circle like plate to the router base
and think about how to still being able to attach the wight/bricks.
Now here is something important:
By creating the schematic I think I found out (but need to calculate it through) that this whole sled reconstruction might be unnecessary in the first place.
By simply attaching the chains to the two points at 120degrees clockwise and counter clockwise, the router will run in the sweet spot of the triangle that the chains form and thus the construction will be mathematically stable. This has the obvious advantage, that the whole construction stays clean and easy to build up.
most of the bearing sets posted are not able to support radial forces, only
axial (and remember, the router needs to poke through at the center, so if the
bearing has a pin there to keep it from flying apart, that won’t work
none of these are enough cheaper than http://www.ebay.com/itm/282350157139 to be
worthwhile.
That is a good point about lazy susans not being designed for radial loading. I would be interested to see a comparison between them and a ring with good bearing pulleys. I would also think that a flat plate cut into a ring shape would provide the best resistance to deformation due to the radial loads applied while also minimizing cost and potentially simplifying attachment to the sled using some sort of standoff screwed to both the sled and the plate ring.
Hi folks. New here. Apologies in advance for total ignorance. I’m impressed by the ingenuity.
Right now, everyone seems to be taking it for granted that the chains should be attached very low, near the base of the router. This requires a ring around the router.
Is it a bad idea to raise the chains (and their motors) in the z-axis and create a pivoting connection point directly above the router’s motor in line with the center of the router bit? In effect, the chains really would be physically connecting to form a triangle.
I can think of objections to controlling the router from so high up (“high” in the z-axis), but do they outweigh the gains? And are they overcome-able?
(I haven’t tested any of this. My lovely Maslow kit just arrived. I won’t be able to assemble and play until I move into a new space in September.)
I think that would be bad. Wouldn’t the router want to rotate around your connection point when moving in the X axis? This rotation would lift the router in the Y axis. The faster you moved in the X direction the more abrupt this would become.
1 Like
Instead of trying to get a perfect ring (which is nof so easy) you start with a 1/4" sfeel plate and cut (laser/cnc mill/waterjet) a ring from it. So it would still look like a hollow circle, but its cross section would be rectangular instead of circular.
Because the cross section has these inner and outer vertical surfaces, it would give the chain pulleys more mechanical advantage to resist the sled pivoting about the x-axis or the y-axis.
Maybe the chains then don’t even need pulleys. They could be attached to nylon-lined, hollow steel segments, where the ring described above fits in these segments inside the nylon lining like a piston fits in a bore or cylinder. Advantage: Low friction, dust resistant,easy to replace nylon lining as it wears (the steel segments split open for this), and able to resist torsion as well as tension.
1 Like
Noting that it may be possible to “simplify” centering and attaching a ring to the bolt holes left by removing router handles, if a relatively small ring is used. Could mean no brackets added to the sled. As the majority of folks are using the Ridgid 2200, and most routers have some sort of handle knobs this should be a pretty general solution.
The first piece of hardware that comes to mind that might make this possible are simple turnbuckles, as I couldn’t think of a bolt I could tighten with the ring attached. With a turnbuckle, one end can be a piece of allthread screwed into the handle holes, the other end is an eyebolt with a diameter eye roughly the size of your rings but doesnt need to be exact. You could also just get a bolt and cut a groove the ring can sit in, or find a bolt with a deep channel maybe? Anyway, adjust the turnbuckles to push out until it holds the ring in place.
If the ring had enough flex, you could forgo the turnbuckles and simply use eyebolts cut to make a sort of saddle, screw them into the handle holes so they are just over the rings inner diameter, then force the ring into the saddles.
1 Like
The wood-ring-plate could as well just be a half-circle,
since the chains would only just run around the lower arc.
Sure some construction is needed to fit such a circle like plate to the router base
and think about how to still being able to attach the wight/bricks.Now here is something important:
By creating the schematic I think I found out (but need to calculate it through) that this whole sled reconstruction might be unnecessary in the first place.
By simply attaching the chains to the two points at 120degrees clockwise and counter clockwise, the router will run in the sweet spot of the triangle that the chains form and thus the construction will be mathematically stable. This has the obvious advantage, that the whole construction stays clean and easy to build up.
Wouldn’t the sled tend to roll to one side or the other if only one chain is being adjusted?
There is now a settings option to switch between what I’m calling “quadrilateral” kinematics (the normal kind) and what I’m calling “triangular” kinematics (where the chains meet at a point or simulate meeting at a point)
The option is very much untested at this point, so use with caution. I’ll test it thoroughly on Monday.
There is also a settings option for the radius of the rotation. If the chains truly do meet at a point, this would be zero.
2 Likes